Disaster recovery is a term used to describe the strategy and processes used by an information technology organization to ensure that in the event of a disaster that damages
Backing up the data of an organization is critically important to help not only with disaster recovery, but also to help to recover from day to day mistakes, such as when a user inadvertently deletes an important document, or to help protect data lost from a malicious intruder. There are many theories and methods of performing data backup. Most commonly, only an application's data is backed up instead of the application itself or the operating system. This is typically done to save space within the backup system (and to reduce the cost of the backup system) because the application's code and the operating system itself changes rarely. It is also traditionally more difficult to restore an operating system and applications from backups versus reinstalling them and then applying data afterwards. Because of these methods, the restoration of systems becomes a much harder, longer task. Although complete system restoration is rare (hopefully), the amount of time it takes to restore systems is critical to business continuance.
Virtualization technology can help improve disaster recovery strategies in many ways. Since virtual machines, their guest operating systems, all installed applications, and data reside within one or more files on the host server, it is simple to backup an entire virtual machine. To backup the virtual machine, make a copy of the virtual machine's configuration files and its virtual disk files. Backing up entire virtual machines usually does require more backup storage space because of the size of the virtual disk fi les, but the benefits are worthwhile. In the event of a disaster, the entire virtual machine can be restored in the time it takes to recopy the virtual machine's configuration and disk files. Installation and configuration of neither operating systems nor applications is necessary. In the same amount of time it takes to rebuild a single standard Windows server, configure it, apply patches and updates, install applications, and restore data, many virtual machines can be restored, possibly 10, 20, 30, or more.
Another aspect of using virtualization in regards to disaster recovery is that when a disaster occurs and a recovery must take place, the proper hardware components must be obtained on which to recover the systems. Traditionally, older hardware is stored off site to be used for recovery and only the core, mission critical servers can be restored because of hardware limitations. Any secondary systems cannot be restored until more hardware is acquired. Using virtualization, a few physical servers able to support the virtualization platform being used are required for recovery. The most critical virtual machines can easily be restored onto any of the virtualization platform-compliant hardware and shortly after that, other secondary systems may also be restored onto the same servers utilizing the unused hardware resources (much like server consolidation).
Virtualization can also help aid in disaster recovery even if not all of an organization's servers are virtualized. For instance, high-performance, highly utilized file and mail servers may not be virtualized in production. Other servers within the organization are consolidated using server virtualization. As part of the organization's disaster recovery strategy, there are virtual machines already created but not used that replicate the functionality of the file and mail servers. When the time comes to recover these servers, the virtual versions provide very quick service restoration times and will usually be adequate until the proper hardware can be acquired.
Use the following table of contents to navigate to chapter excerpts, or click here to view Business cases for server virtualization in its entirety.
|ABOUT THE BOOK:|
|Advanced Server Virtualization focuses on the core knowledge needed to evaluate, implement and maintain an environment that is using server virtualization. It emphasizes the design, implementation and management of server virtualization from both a technical and a consultative point of view. It provides practical guides and examples, demonstrating how to properly size and evaluate virtualization technologies. This volume is not based upon theory, but instead on real-world experience in the implementation and management of large-scale projects and environments. Currently, there are few experts in this relatively new field, making this book a valuable resource. Purchase the book from Amazon|
|ABOUT THE AUTHORS:|
|David Marshall is currently employed as a software engineer for Surgient Inc., a software company based in Austin, Texas, that provides software solutions that leverage x86 server virtualization technologies. He holds a B.S. in finance and an Information Technology Certification from the University of New Orleans. He has been working with virtualization software for the past six years. Dave McCrory works as chief scientist for Surgient Inc. He has filed several patents around server virtualization and management of virtual machines and has worked with virtualization technology for more than five years. Wade A. Reynolds is employed as a senior consultant by Surgient Inc. He has been designing and implementing enterprise solutions based on virtualization technology on a daily basis for more than three years, including VMware ESX Server and Mircosoft Virtual Server from its pre-beta release.|
This was first published in January 2008